Towards Unsupervised Whole-Object Segmentation: Combining Automated Matting with Boundary Detection

Abstract

We propose a novel step toward the unsupervised segmentation of whole objects by combining "hints" of partial scene segmentation offered by multiple soft, binary mattes. These mattes are implied by a set of hypothesized object boundary fragments in the scene. Rather than trying to find or de?ne a single "best" segmentation, we generate multiple segmentations of an image. This re?ects contemporary methods for unsupervised object discovery from groups of images, and it allows us to de?ne intuitive evaluation metrics for our sets of segmentations based on the accurate and parsimonious delineation of scene objects. Our proposed approach builds on recent advances in spectral clustering, image matting, and boundary detection. It is demonstrated qualitatively and quantitatively on a dataset of scenes and is suitable for current work in unsupervised object discovery without top-down knowledge.